Open-source alternatives to CoreMLExample

This is a PyTorch object detection framework that implements the Single Shot MultiBox Detector for identifying and localizing multiple objects within images and video. The project provides a neural network architecture designed for single-shot object detection, which predicts bounding boxes and class labels in one pass. The implementation includes a real-time object detector capable of processing live video streams to track and label objects across sequential frames. It also features a complete computer vision training pipeline for preparing image datasets and training model weights. The fra

TCS humAIn This is a Flutter application that is used to locate the license plate out of a picture given to the application. Cue the Drum Rolls for what I am about to disclose. With the help of Sayak Paul the tensorflow model that was 255mb was cut short to a 2mb file. TFLite did the trick for…

Generate image analogies using neural matching and blending.

A PyTorch implementation of DenseNet.

Classification with PyTorch.

Convolutional Recurrent Neural Network

Pixel-wise segmentation on the VOC2012dataset dataset using pytorchpytorch.

Base pretrained models and datasets in pytorch (MNIST, SVHN, CIFAR10, CIFAR100, STL10, AlexNet, VGG16, VGG19, ResNet, Inception, SqueezeNet)

Compressed Video Action Recognition

PyTorch implementation of PNASNet-5 on ImageNet

A demo for iOS machine learning framework : Core ML

This is a Gluon implementation of MobileNetV2 architecture as described in the paper Inverted Residuals and Linear Bottlenecks: Mobile Networks for Classification, Detection and Segmentation.

3DDFA is a 3D face reconstruction tool that generates three-dimensional facial meshes and 68 structural landmarks from a single two-dimensional input image. The project provides utilities for estimating facial pose and depth maps to determine the orientation and position of a face. It includes a geometry exporter to save reconstructed facial shapes into standard file formats for use in external 3D modeling software. The software further covers automated face cropping using landmark-based and landmark-free identification methods, as well as the extraction of specialized mathematical represent

OpenPose is a real-time pose estimation engine designed to detect and track human body, face, hand, and foot landmarks. It functions as a multi-person motion tracker, identifying the spatial coordinates of multiple individuals simultaneously within video streams or static images. Beyond two-dimensional detection, the software acts as a three-dimensional kinematics processor, reconstructing spatial movement data from single or multiple synchronized camera perspectives. The system distinguishes itself through a bottom-up approach that utilizes part-affinity fields to associate body parts across

👀 iOS11 demo application for age and gender classification of facial images.

🌸 iOS11 demo application for flower classification.

🖼 iOS11 demo application for image style classification.

Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Neural Networks

This is the official implementation of Cross-domain Compositing (CDC), a local, inference-time, image editing method which utilizes pretrained diffusion models for image compositing in various domains. We base our method on previous work in global image editing in inference-time, and propose a…

PyTorch implementation of Deformable Convolution